0
$\begingroup$

I have seen similar questions regarding this issue, but this seems a little odd. Whenever a Boolean Modifier is used between two components, everything disappears. Although whenever I scale both components down half the size, the modifier works perfectly normal. The union turns out exactly as anticipated, which I could then scale back up to the normal size. Having to do this work around seems a little odd, since its telling me there are no issues with the geometry of the components.

I did complete a basic list of checks on the components before any boolean modifier is used in order to limit any issues that may occur:

  • Applied any scaling translations on the components
  • Removed any doubles
  • Recalculated all the normals making sure each is facing outward
  • Verified components are manifold to the best of my knowledge

Is it possible, due to a larger volume being taken up, that blender ended up running out of memory while executing the modifier? This would make sense since the values used during the calculations would end up larger based on the coordinates each point has (everything is farther apart). Any thoughts?

I should note that 4GB of RAM is available and I'm currently working with version 2.75 of Blender. I have also attached the .blend file of the two components I am trying to merge together for anyone who feels the need to try the union on their system.

$\endgroup$
2
  • $\begingroup$ If you use the triangulate command on the larger mesh, the modifier works. Not sure if that's a workaround that you can live with, but it works. I think that's there's some kind of strange tolerance error going on but it's hard to prove. $\endgroup$
    – SamFall
    May 25, 2016 at 0:46
  • $\begingroup$ Maybe it is some tolerance issue. It seems like blender may think the two objects are farther apart and not interfering with one another. I believe the same disappearance is desired whenever, say two balls are not touching. Although, clearly the two objects in the .blender are touching and should interact in some fashion. $\endgroup$
    – Jake
    May 25, 2016 at 16:19

1 Answer 1

1
$\begingroup$

No, Blender did not run out of memory. If it had, Blender would have crashed. Furthermore, the scale (applied or not) of an object has nothing to do with it's complexity. A mesh with 5000 polygons takes up exactly (within a few bits) the same RAM no matter what scale it is.

This is probably a bug. If you can reproduce it reliably, you should look for it in the bug tracker and if it hasn't already been reported, report it.

$\endgroup$
2
  • $\begingroup$ I can confirm that it is still happening in 2.77,temporarily scaling down both objects does correctly preform the Boolean operation, so it does sound like a bug as you mentioned, so probably worth reporting to the bug tracker. $\endgroup$ May 24, 2016 at 17:10
  • $\begingroup$ I'll see what I can look into on the bug tracker and report the issue if there isn't anything there. That boolean modifier always seems to cause issues it seems. Thanks for the inputs. $\endgroup$
    – Jake
    May 25, 2016 at 16:15

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .